Two cohort of patients will be followed to examine the course of illness in Alzheimer's disease. We will contrast two hypotheses of illness progression: 1) initiation in the medial temporal lobe and spread through the corticocortical projections to the lateral temporal, frontal and parietal lobes and 2) multilocal or global initiation as the variable regional expression of a common underlying diffuse pathological process. In the first cohort (data collected at the University of California, Irvine), 39 patients with mild to moderate Alzheimer's disease received PET Scans and co-registered MRI Scans, neurological examinations, and a battery of neuropsychological assessments in 1989-1990. These patients and their matched elderly normal controls are currently receiving 2-year clinical followup, and will be contacted regularly during the course of the project. These patients have already received a repeat scan. For the second cohort (data to be collected at Mt. Sinai), the ADRC Clinical Research Support Core will recruit and diagnose patients from the Mt.Sinai geriatric programs. We will PET and MRI scan and assess 16 patients each year, eight with memory deficits not meeting NINCDS criteria for AD (Mini-mental more 20 and CDR=0.5; termed "questionable AD") and eight meeting criteria for mild Alzheimer's disease (CDR 0.5 or 1.0) and meet NINCDS ADRDA criteria for AD. Patients will receive a second set of scans after 24 and 48 months; during this funding period not all patients will receive 48 month followup. The ADRC Core will provide followup at six month intervals. Approximately 68% of patients entering the ADRC will come to autopsy. While relatively few will be available in the first five year period, we will have the PET and MRI database for analysis of regional neurofibrillary tangle and plaque counts for the second grant period. From this sample rescanned at regular intervals we will learn if medial temporal lobe change is systematically followed by lateral cortical deficits. The specificity of neuroanatomical sites identified in cohort 1 will be replicated at higher resolution in cohort 2. A complete life span normal control group (ages 20-80) performing the same task during FDG uptake will be available for comparison. Scans will be carried out on our new high-resolution, head dedicated GE 2048 PET Scanner with FDG as the tracer and a verbal memory task. Matching MRI Scans will be obtained for co-registration and assessment of gray/white matter change.